THE ECONOMICS OF TOBACCO TAXATION AND EMPLOYMENT IN INDONESIA - Health, Population, and Nutrition Global Practice - World Bank Document
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized
THE
Global Practice
Health, Population, and Nutrition
ECONOMICS
OF TOBACCO
EMPLOYMENT
IN INDONESIA
TAXATION AND
Cover photos (clockwise from left): Hand holding green tobacco leaf. Photo by bdspn / iStock. Smiling Asian lady. Photo by joakimbkk / iStock. Gateway on the road to oil palm plantations. Photo by Flore de Preneuf / World Bank. Man in crop of tobacco plants. Photo by ribeiroantonio / iStock.
THE ECONOMICS OF TOBACCO TAXATION AND EMPLOYMENT IN INDONESIA Health, Population, and Nutrition Global Practice
SMOKING HAS BEEN A MAJOR CONTRIBUTOR TO THE DISEASE BUR IN INDONESIA. SMOKING PREVALENCE AM WORKING-AGE INDIVIDUALS HA EXCEEDED 30
LIST OF ACRONYMS BPJS Badan Penyelenggara Jaminan Sosial FGD Focus Group Discussion FT Full-time FTE Full-time Equivalent GDP Gross Domestic Product KIS Kartu Indonesia Sehat KKS Kartu Keluarga Sejahtera KPS Kartu Perlindungan Sosial NCD Noncommunicable Diseases PBI Penerima Bantuan Iuran PPP Purchasing Power Parity Rp Indonesian Rupiah SKM Machine-made kretek SKT Hand-made kretek SPM White cigarettes
EXECUTIVE SUMMARY Indonesia has one of the most complex cigarette excise tax structures in the world. The current cigarette excise tax is divided into 12 tiers, which are based on manufacturers’ type of cigarettes, the number of cigarette productions, and per-unit retail price. The tiers in the cigarette excise tax structure aim to accommodate small-scale cigarette firms, espe- cially hand-rolled kreteks (SKT) firms. The rationale for such a structure is that smaller SKT firms account for more than half of the total factories in the tobacco industry. Moreover, these firms are responsible for employing a significant share of the workers in tobacco manufacturing. Such a system favors downward substitution to lower priced products and has a limited impact on smoking reduction. This report aims to contribute to the policy debate over the tobacco excise tax reform and, specifically, any effects it might have on employment in Indonesia. It uses data from the Central Bureau of Statistics to observe the trends in employment and output in the tobacco sector and estimate the potential impact on tobacco employment (gross effect) from raising cigarette taxes in Indonesia. The report shows that tobacco manufacturing represents only a small share of economy-wide employment (0.60 percent). Additionally, the contribution of tobacco manufacturing to employment in the manufacturing sector is quite small (5.3 percent) in comparison to the contribution of the food (27.43 percent), garment (11.43 percent), and textile (7.90 percent) sectors. Although small, tobacco manufacturing jobs are heavily con- centrated, with about 94 percent of tobacco manufacturing workers in Central Java, East Java, and West Nusa Tenggara. In these provinces, several districts are quite dependent on tobacco sector employment, for example, Kudus (30 percent), Temanggung (27.6 per- cent), and Kediri (26 percent). We found that most tobacco manufacturing workers are females and unskilled. The share of female workers in tobacco manufacturing is 66 percent, which is the highest in the manufacturing industry. Other sectors in which most workers are female are the garment sector (63 percent) and textiles sector (56 percent). About 69 percent of total workers in the tobacco sector completed at most, junior high school. Among tobacco households, the share of tobacco households with female tobacco workers is 78 percent. Average years of schooling completed by tobacco manufacturing workers is among the lowest at 8.22 years, which is comparable to the average years of schooling completed by workers in the food processing sector (8.50 years), manufacturing of wood products sec- tor (7.31), nonmetallic metal products sector (7.42), furniture sector (8.77), and recycling sector (5.20). Tobacco households are partially dependent on income from tobacco employment, which represents 60 percent of household income, on average. Additionally, among these households, only 9 percent have a female as the primary income earner. These find- 4 // Executive Summary
ings from secondary data (the Sakernas data) are consistent with the ones from the World
Bank/ACS kretek workers survey in Indonesia—the proportion of wage income from kretek
is 54 percent and the share of female tobacco workers as primary earner is about 10 per-
cent (World Bank, 2017a).
Productivity of tobacco sector workers, measured by the output per worker, is rela-
tively low in comparison to productivity of workers in the comparable sector. A typi-
cal worker in the medium and large tobacco firms can produce IDR 104 million (US$7,761)
worth of products annually. For comparisons, a typical worker in the food and drink, and
textile industries can produce IDR 265 million (US$19,776) and IDR 300 million (US$22,388)
worth of products respectively.
The estimated output elasticity of labor demand for the cigarette manufacturing
sector is 0.160. This means that a 1 percent decrease in output results in a 0.16 percent
decrease in employment in the cigarette manufacturing sector. On the other hand, the
estimate for the tobacco processing sector implies that a 1 percent decrease in output
corresponds to a 0.092 percent decrease in employment in the tobacco processing sector.
We predict that raising cigarette taxes by an average of 47% and simplifying the
cigarette tax structure to 6 tiers will reduce cigarette demand by 2 percent, increase
government revenue by 6.4 percent, and reduce gross employment in tobacco
manufacturing sector by less than 0.50 percent. That means that a reduction of 2,914
tobacco manufacturing jobs, most of them in the SKT industry (2,245 less jobs). Given the
additional revenues government will obtain with the reform (IDR 10,915 billion), there is
scope to implement measures to reduce the impact on the tobacco workers’ livelihoods
(such as cash transfers or expanded access to social safety nets) or to find alternative occu-
pations for the workers affected (retraining programs, educational grants, etc.).
The evidence presented in this report clearly shows that the gross employment
impact of reforming tobacco excise taxes and structure in Indonesia is not as big as
previously thought. Analyses presented here are complemented by other evidence pre-
sented in the World Bank/ACS Indonesia Tobacco Studies, which highlighted the economic
and social costs of tobacco workers’ and farmers’ livelihoods. Given the additional reve-
nues the government will obtain with the reform (IDR 10,915 trillion), there is an oppor-
tunity to implement measures to reduce the impact on the tobacco workers’ livelihoods
(such as cash transfers or expanded access to social safety nets) or to find alternative occu-
pations for the workers affected (retraining programs, educational grants, etc.).
It is important to note that these estimates represent the gross employment effect
of lower cigarette consumption. When prices of cigarettes increase, consumers may
shift their consumption to other goods and services which will create jobs in these sec-
tors. Evidence has shown that the job losses in the tobacco sector (gross effect) are usually
compensated with job creation in the other sectors (net effect). For Indonesia, Ahsan and
Wiyono (2007) estimated positive effects varying from 84,340 to 281.153 jobs with tax
increases of 25 percent and 100 percent, respectively (Ahsan and Wiyono, 2007).
5• For kretek hand-rollers
o First, the most vulnerable groups in the affected population who would need
immediate income support in the event of job loss include the workers who are less
educated, older, heads of their households, and who contribute a significant pro-
portion of total household income from kretek rolling. The government can provide
income support to these workers with less than 2% of the revenue gained from a tax
increase; and
o The government (Ministry of Finance and Ministry of Social Affairs) should provide
temporary income support using the existing Social Assistance programs (such as
the unconditional cash transfer program, Bantuan Langsung Sementara Masyarakat
- BLSM) and identify alternative employment or income-generating opportunities in
the affected regions. The re-training of laid-off kretek workers should be designed to
accommodate transitions into these alternative employment opportunities.
• For (tobacco and clove) farmers:
o The government should help to improve supply chains and value chains for
other goods in tobacco-growing areas. Many former tobacco farmers are making
a better living growing other common, locally grown crops (e.g., corn, sweet potato,
and green vegetables), an outcome that could be further enhanced with even small
investments by governments in improved supply chains for these products. Results
from the World Bank/ACS survey suggest that current tobacco farmers are already
growing many of these crops, so it is an issue of shifting their factors of production to
maximize economic opportunity.
o The government should help to facilitate access to credit for tobacco
farmers. Greater access to capital through improved credit schemes could help to
improve the possibilities for tobacco farmers to cultivate other crops and/or develop
other nonagricultural economic enterprises. That could be in the form of grants or
low-interest loans to farmers willing to move away from tobacco cultivation; and
o Specifically, on clove farmers, it is important to emphasize that clove
farming is not particularly profitable for most clove-producing households in
at least a couple of major clove-producing districts. The government needs to
research which alternatives could be viable and target the least profitable areas for
switching and help these farmers make successful transitions to growing other crops
and/or economic activities.
6 // Executive SummaryACKNOWLEDGMENTS
The preparation of this report was carried out under the World
Bank Global Tobacco Control Program coordinated by Patricio
V. Marquez, with the support of the Bill & Melinda Gates
Foundation and the Bloomberg Foundation.
The report is part of the Indonesia Tobacco Employment Studies
and was prepared by a team comprised of:
Gumilang Aryo Sahadewo (Research Faculty Universitas Gadjah Mada
and Consultant World Bank);
Roberto Iglesias (Director, Center of Studies on Integration and
Development, CINDES, Brasil);
Edson C. Araujo (Senior Economist and Task Team Leader, World Bank);
Nigar Nargis, PH.D. | Scientific Director, Economic and Health
Policy Research
Pandu Harimurti (Senior Health Specialist and Task Team Leader,
World Bank);
Jeffrey Drope (Scientific Vice-President, American Cancer Society)
Qing Li (Senior Data Analyst, American Cancer Society);
Josefine Durazo (Survey Specialist, World Bank);
Firman Witoelar (Director of Research, SurveyMeter); and
Bondan Sikoki (Chairperson, SurveyMeter).
The report greatly benefited from comments, inputs, and advice provided
by Violeta Vulovic (Economist, World Bank), Frank Chaloupka (Professor,
Institute for Health Research and Policy, University of Illinois, Chicago),
Jeremias N. Paul (Coordinator of Tobacco Control, WHO) and Michael Weber
(Senior Economist, World Bank).
7TABLE OF CONTENTS
List of Acronyms 3
Executive Summary 4
Acknowledgments 7
List of Tables 9
List of Figures 9
1. Introduction 11
2. The Economics of Tobacco Taxation 15
2.1 — Tobacco Consumption, Health Outcomes, and Economic Costs 15
2.2 — Tobacco Taxes and Tobacco Consumption 17
2.3 — Tobacco Taxation and Tax Revenues 18
2.4 — Employment Impacts of Tobacco Taxation 19
3. Cigarette Tax Policies in Indonesia 27
3.1 — Effects of Cigarette Consumption, Revenues, and Employment 29
4. Employment in the Indonesian Tobacco Sector 35
4.1 — Data Sources 35
4.2 — 23 Employment in the Indonesia Tobacco Farming Sector 36
4.3 — Employment in the Tobacco Manufacturing Sector 38
5. Simulations on the Effects of Raising Cigarette Taxes on Employment 51
6. Conclusion 61
References 63
Annex I. Estimation of the Wage Equation 67
Annex II. Sample Selection Procedure for the Manufacturing Industry Data 68
Annex III. Estimation of Output Elasticity of Labor Demand 70
Annex IV. Estimation of Price Elasticity of Demand 72
Annex V. Simulation of the Effects of Raising Cigarette Prices on Employment 74
8 // Table of ContentsList of Figures
Figure 1. Trends of the Cigarette Excise Tax in Indonesia in real terms, 2010–2017 29
Figure 2. Revenue from the tobacco excise, 2005–2015 31
Figure 3. Trends of cigarette production in Indonesia, 2010–2016 32
Figure 4. Number of tobacco farmers and share of tobacco farmers to total workers, 1990–2011 37
Figure 5: Number of medium and large tobacco manufacturers, 2000–2014 40
Figure 6. Share of tobacco employment to total manufacturing employment, 2014 41
Figure 7. The concentration of tobacco manufacturing workers, 2014 42
Figure 8. Number of workers per firm by industry 43
Figure 9. Productivity of workers, 2000–2014 47
Figure 10. Relationship between output and labor in medium and large industry, 2000–2014 47
List of Tables
Table 1. Selected studies on net employment impact of tobacco control policies 21
Table 2. Type of external tobacco trade and likely net employment impacts 22
Table 3. Cigarette excise tax in Indonesia, 2017 28
Table 4. Distribution of tobacco farmers across regions, 2014 38
Table 5. Distribution of clove farmers across regions, 2014 39
Table 6. Employment structure of the Indonesian tobacco industry, 2014 40
Table 7. Concentration of workers in selected districts, 2014 43
Table 8. Characteristics of tobacco sector workers, 2011, 2013, and 2015 44
Table 9. Estimation results of the wage equation, 2001–2015 46
Table 10. Estimation of own-wage and output elasticity of labor demand, 2000–2014 48
Table 11. Analysis of the 2018 Cigarette Excise Tax Regulation 52
Table 12. Simulations on the effects of the 2018 Tax Law on excise revenue 53
Table 13. Estimated number of workers in each tax tier 54
Table 14. A hypothetical government scenario 56
Table 15. The effects of raising taxes on revenue: a hypothetical government scenario 57
Table 16. The effects of raising taxes on employment: a hypothetical government scenario 58
Table 17. The effects of raising taxes on employment: World Bank and ACS scenario 59
Table 18. Summary statistics, National Labor Force Survey Sample 67
Table 19: Selected sample, Annual Survey of Manufacturing Industry 68
Table 20: Summary statistics, Annual Survey of Manufacturing Industry 71
Table 21: Estimation of price elasticity of cigarette demand, 2015 73
9SMOKING HAS The Economics of Tobacco Taxation and Employment in Indonesia BEEN A MAJOR CONTRIBUTOR TO THE DISEASE BUR IN INDONESIA. SMOKING PREVALENCE AM WORKING-AGE INDIVIDUALS HA EXCEEDED 30 10 // Table of Contents
1
INTRODUCTION
Smoking has been a major contributor to the disease burden in Indonesia. Smoking
prevalence among working-age individuals has exceeded 30 percent since 2001. In 2013,
smoking prevalence among males was 66%, while among females was 6.7%. Smoking
prevalence among children ages 10–14 in 2013 was 3.7%, twelve times higher than in
1995 (IAKMI, 2014; Ahsan, 2015). Diseases attributable to smoking include hypertension,
acute respiratory infection, coronary heart disease, cardiovascular diseases, selected can-
cers, and perinatal disorders (IAKMI, 2014; Kosen et al., 2012; Kristina et al., 2015). In 2013,
healthy years lost at the population level due to smoking-induced diseases was estimated
to be 6.2 million disability-adjusted life years or DALY (IAKMI, 2014).
The government of Indonesia continues its efforts to reform the cigarette excise
tax system. The main objectives of the excise tax reform are to reduce smoking preva-
lence and to increase tax revenues. Studies suggest that a 10 percent increase in cigarette
excise tax would lead to a reduction in cigarette consumption by 0.9 to 3 percent and an
increase in government revenue by 6.7 to 9 percent (Hidayat and Thabrany, 2010; Setyo-
naluri et al., 2008). Currently Indonesia has one of the most complex cigarette tax struc-
tures in the world, which favors downward substitution to lower priced cigarettes (World
Bank, 2015). Cigarette prices across all tiers increased at a modest rate between 2010 and
2017 as the government continued to increase cigarette taxes; nevertheless cigarette
prices were more affordable in recent years than they were in 2000 owing to a faster
income growth (NCI-WHO, 2017). The main argument against adopting such a complex
cigarette tax structure was to protect employment by differentiating firms with different
production scales with those that employ more workers, such as hand-rolled kreteks (SKT)
workers.
Despite the concerns, tobacco manufacturing represents only a small share of
the economy-wide employment (0.60 percent). Additionally, the contribution of the
tobacco manufacturing to employment in the manufacturing sector is quite small (5.3
percent) in comparison to the contribution of the food (27.43 percent), garment (11.43
percent), and textile (7.90 percent) sectors. Although small, tobacco manufacturing jobs
are heavily concentrated in Central Java, East Java, and West Nusa Tenggara, where about
94% of tobacco manufacturing workers are employed. In these provinces, several districts
are quite dependent on sector employment, as for example Kudus (30 percent), Temang-
gung (27.6 percent), and Kediri (26 percent). A tax change shock that affected the tobacco
sector would affect these districts most.
11The Economics of Tobacco Taxation and Employment in Indonesia The objective of this report is to analyze the recent employment trends in the Indonesian tobacco industry and estimate the potential effects of raising cigarette taxes on employment in the tobacco manufacturing sector. The report provides new evidence to contribute to the ongoing debate about the effects of raising cigarette taxes on tobacco sector employment. It complements the current analytical work conducted by the World Bank, in partnership with the American Cancer Society, to explore the employment conditions and livelihoods of tobacco and clove farmers and kretek rollers in Indonesia. The report is part of the World Bank technical assistance to the government of Indonesia in the areas of revenue policy reform and health systems reform. This report is structured as follows: Section two provides a review of the global evidence on the impacts of raising cigarette taxes on population health outcomes and on the economy. It briefly reviews the economic issues around tobacco taxation, including a summary of the global evidence on the effects of taxation on smoking reduc- tion, fiscal revenues, tobacco production and employment in the sector. Section three presents an overview of the cigarette tax reforms in Indonesia in recent years and dis- cusses the government of Indonesia’s plans for reforming the cigarette tax structure. Sec- tion four discusses the employment trends in the tobacco industry in Indonesia, analyzes workers’ characteristics and compares workers to similar sectors and socio-demographic profiles. The section also discusses the potential impacts of raising cigarette taxes on employment by presenting results of simulations. The final section, section five, discusses the results in light of the current debate over cigarette tax reform in Indonesia and pro- vides policy recommendations on the employment aspects of the reform. 12 // Introduction
13
SMOKING HAS BEEN A MAJOR CONTRIBUTOR TO THE DISEASE BUR IN INDONESIA. SMOKING PREVALENCE AM WORKING-AGE INDIVIDUALS HA EXCEEDED 30
2
THE ECONOMICS OF
TOBACCO TAXATION
Raising cigarette prices through taxes is a cost-effective way to reduce cigarette
consumption as recommended by the World Health Organization Framework Con-
vention on Tobacco Control (WHO FCTC, 2003). The objective of raising cigarette taxes
is to decrease consumption of cigarettes through higher prices.1 Additionally, raising cig-
arette taxes can increase government’s tax revenues, which can be allocated to finance
complementary tobacco control policies and social investments such as health and
education. Experiences in both high and low-and middle-income countries (LIMCs) show
how tobacco tax revenues can be used to expand health care coverage. In France, for
example, tobacco tax revenues are used to supplement the funding for public health ser-
vices. In Philippines, higher tobacco and alcohol tax revenues have been used to expand
health coverage for the poor. On the other hand, the tobacco industry argues that lower
consumption of tobacco products due to higher tobacco tax can have negative effects
on production and, consequently, on employment. This section presents a summary of
the global evidence on the impacts of tobacco consumption on population’s health and
the impacts of tobacco taxation on fiscal revenues and employment.
2.1 Tobacco Consumption, Health Outcomes, and Economic Costs
Smoking is a major cause of morbidity and premature mortality. Tobacco consump-
tion has been directly linked to diseases of the circulatory system (e.g., ischemic heart
and cerebrovascular diseases); cancers of the trachea, bronchus and lung, esophagus,
oropharynx, larynx, stomach, liver, pancreas, kidney and ureter, cervix, bladder, colon/
rectum, as well as acute myeloid leukemia; chronic respiratory diseases (e.g., asthma,
chronic obstructive pulmonary disease); and metabolic diseases such as diabetes mel-
litus (U.S. Department of Health and Human Services, 2014). Besides the harm of direct
consumption, it has been proven that secondhand smoking (sometimes referred to as
passive smoking, environmental tobacco smoke, or tobacco smoke pollution) also has
damaging health consequences. Secondhand smoking is quite dangerous because there
are at least 50 carcinogenic chemicals inhaled by those who are around smokers, and the
1 We use the term cigarette taxes and tobacco taxes interchangeably because cigarettes are a major form of tobacco product in
Indonesia.
15The Economics of Tobacco Taxation and Employment in Indonesia scientific evidence shows that there is no safe level of exposure to secondhand smoking (US NCI-WHO, 2016). The World Health Organization (WHO) estimates that approximately 7.2 million deaths per year worldwide are attributable to smoking, more than 6 million of those deaths are the result of direct tobacco use, while around 900 000 are the result of non-smokers being exposed to second-hand smoke (WHO, 2017a). Half of these deaths occurred in LMICs. By 2030, the annual death toll could reach 10 million if no tobacco control measures are taken (WHO, 2008). Smoking is one of the major risk factors for noncommunicable diseases (NCD) deaths. NCDs kill 40 million people each year, which is equivalent to 70% of all deaths globally. Each year, 15 million people die from a NCD between the ages of 30 and 69 years; over 80% of these premature deaths occur in LMICs (WHO, 2017b). Worldwide, approximately 14% of adult deaths from NCDs are attributed to tobacco use, includ- ing 10% of all adult deaths from cardiovascular diseases (14% among men, 6% among women), and 22% of all adult deaths from cancer (32% among men, 11% among women). The clear majority (71%) of adult lung cancer deaths (78% among men, 53% among women) were attributable to tobacco. In addition, 36% of all adult deaths from diseases of the respiratory system were attributable to tobacco (42% among men, 29% among women) (WHO, 2012). Tobacco smoking is also an important risk factor for chronic obstructive pulmonary disease (COPD). In 2004, about 49% of the COPD deaths among adult men and 34% of COPD deaths among adult women were attributable to tobacco (WHO, 2012). NCDs have become a major public health concern in Indonesia. The WHO estimates that the proportional mortality due to NCDs has increased from 50.7% in 2004 to 71% in 2014 (WHO, 2014). Tobacco smoke is the fourth risk factor that contributes to most death and disability combined from NCDs, after dietary risks, high blood pressure, and high fast- ing plasma glucose (IHME, 2017). In 2012, NCDs accounted for more disability-adjusted life years (DALYs) than communicable diseases—approximately 476 million and 240 mil- lion DALYs, respectively (WHO, 2014). Smoking is also responsible for deaths from communicable diseases. Approximately 5% of global deaths from communicable diseases are attributed to tobacco, including 7% of all deaths due to tuberculosis (TB) and 12% of deaths due to lower respiratory infections (WHO, 2012). A systematic review of the literature found a significant positive relationship between exposure (passive or active) to tobacco smoke and TB infection and disease, independent of various potential health issues including alcohol use and socio- economic status (Salma et al., 2007). Recurrent TB and mortality resulting from TB were also associated with active smoking. This relationship is particularly important in Indo- nesia, given that the country has one of the highest TB infection rates in the world and where tuberculosis is one of the top causes of death (USAID, 2009). 16 // The Economics of Tobacco Taxation
The risk of death due to smoking declines with the length of time from an individ-
ual quits smoking. In other words, the sooner one quits smoking, the longer she/he will
live. Studies have been conducted for the United States (Burns et al., 1997), India (Jha et
al., 2008), and Germany (Neubauer et al., 2006) that have provided a quantitative relation-
ship between the length of smoking, smoking cessation, and benefits of quitting smok-
ing concerning mortality, showing that the reductions in relative risk of heart disease and
stroke are more immediate than the effects on respiratory disease and cancer. Due to the
lengthy time lags for the development of cancers and chronic respiratory diseases asso-
ciated with tobacco smoking, deaths from these illnesses in LMICs may continue to rise,
even if smoking prevalence remains the same or decreases (US NCI and WHO, 2016).
Smoking imposes a substantial economic burden on countries due to increased
health care costs and worker’s productivity losses. Goodchild and colleagues (2016)
measured the economic costs of smoking in 152 countries, representing 97% of the
worldwide smoking population. They considered direct costs related to health care
treatment and indirect costs from productivity losses due to tobacco-related premature
mortality and morbidity. The estimated total economic cost of smoking was equivalent in
magnitude to 1.8% of the world’s annual gross domestic product (GDP) in 2012 and 40%
in LMICs. Of those, 76% are indirect costs related to productivity losses due to morbidity
(35%) and premature mortality (65%). The direct health care costs were estimated at pur-
chasing power parity (PPP) $467 billion (PPP international dollars), which is equivalent to
5.7% of global total health expenditures. The findings from this study highlight the urgent
need to implement comprehensive tobacco control measures to reduce the economic
costs of smoking.
2.2 Tobacco Taxes and Tobacco Consumption
Increasing tobacco and cigarette taxes reduces consumption and, consequently,
can reduce smoking-attributable mortality and morbidity. The impact of tobacco
taxation on the reduction of mortality depends on (i) the magnitude of the price increase
resulting from a tobacco tax increase, (ii) the reaction of consumers to price changes,
that is the price elasticity of demand,2 which is related to smoking behavior (initiating,
reducing intensity, or quitting), and (iii) the relationship between mortality and quitting
smoking. The price of tobacco products in relation to income, i.e. affordability, also mat-
ters on initiation decisions, intensity, or quitting.3 Empirical studies in high- and low- and
middle-income countries have found a negative relationship between cigarette prices
2 Technically, the price elasticity of demand is the percentage change in the consumption of a product in response to a 1% change in
the price of the product, with all else remaining constant.
3 To be effective in reducing tobacco demand, tax and price increases need to be significant to counteract the effect of income
growth on tobacco demand and reduce affordability.
17The Economics of Tobacco Taxation and Employment in Indonesia and smoking. Once prices increase, smokers adjust their decision of consuming through quitting or smoking reduction. Additionally, higher prices, due to a higher tax, act as a deter- rent for new smokers, particularly among the youth or the poor. In high-income countries (HICs), price elasticity estimates are clustered around –0.4 percent (IARC, 2011). In LMICs also show a negative price-elasticities of tobacco demand, ranging from –0.1 to –1.0, with estimates clustered around –0.5 percent (John et al., 2010; Jha and Chaloupka, 2012; Szabo et al., 2016). In other words, in HICs, a 10 percent increase in the price of tobacco is expected to decrease tobacco consumption by 4 percent. In LMICs, a 10 percent increase in price would be expected to decrease tobacco consumption by 5 percent (IARC, 2011). A recent global simulation study shows that an 80 percent increase in excise per pack may lead to 42% increase in price, reduce global annual cigarette consumption by 18 percent and global smoking prevalence by 9 percent (Goodchild, Perucic, and Nargis, 2016). Studies also show that smoking reduction after a tax increase has positive impacts on individual and public health. Several studies show the health benefits of quitting smoking due to tax increases (John et al., 2010; Blakely et al., 2015; Goodchild, Perucic and Nargis, 2016; Szabo et al., 2016). A recent systemic review of studies, published in English between 2000 and 2012, also concludes that tobacco taxation is a highly cost-effective policy, because the costs of intervention are minimal and significantly save health care costs after the tax increase implementation, although many health benefits of quitting take time to materialize (Contreary et al., 2015). 2.3 Tobacco Taxation and Tax Revenues Tobacco and cigarette excise taxation can also be efficient sources of fiscal reve- nue. Given that tobacco demand is relatively inelastic, due to consumer addiction and the lack of close substitutes, tobacco taxes can generate considerable amounts of tax rev- enues, particularly if sales are large. Tobacco taxes may also create fewer distortions in the markets than would result from taxes on goods and services with more elastic demand. Also, given the small number of producers, tobacco taxes are relatively easy to collect at low administration and enforcement cost, as compared to general consumption and income taxes. Experiences in numerous countries indicate that an increase in tobacco taxes will increase nominal (as well as real) tax revenues in the short to medium terms. The magnitude of tobacco and cigarette tax revenue that a government can gen- erate largely depends on the tax system and the demand characteristics. Given a certain income level, per capita or total, a country’s tobacco excise revenue depends on: (i) the level of taxation per unit of tobacco product (either as percent of the price or abso- lute amount of tax per pack); (ii) the number of different tax tiers; (iii) the price elasticity of tobacco demand; and (iv) the volume of tobacco sales. Normally, low levels of tobacco excise revenues are associated with low levels of taxation per unit of product (Chaloupka 18 // The Economics of Tobacco Taxation
et al., 2012). Different tax tiers, such as in Indonesia, allow producers to reduce their tax
burden, expanding production through less taxed products. Also, smokers could avoid
higher rates by switching to lower taxed cigarettes. The result may be lower revenues
than potentially anticipated. The more inelastic the demand, the more government rev-
enues can be generated with a certain tax rate increase. With an inelastic demand, the
proportional reduction in cigarettes purchased by the consumer after the tax increase is
smaller than the proportional increase in tax revenue. As the price elasticity of demand
increases in absolute value, the possibility to raise revenues for a given tax rate change
decreases. Finally, the volume of tobacco sales will determine the possibility of revenue
expansion, given certain demand elasticity and percent changes on the taxation level.
Tobacco taxes are likely to remain high after a significant tax increase, even with
a considerable decline in tobacco use. Chaloupka et al. (2012) argue that over time,
inflation will erode the value of tobacco tax revenues, unless those taxes are increased
often enough to keep pace with inflation. Similarly, as tobacco use declines in response
to other tobacco control efforts, revenues from tobacco taxes will also decline, unless
taxes are increased periodically. Nevertheless, it is possible that tax revenues may remain
higher many years after a significant tax increase than they were before, even in the wake
of a considerable decline in tobacco use. For example, in the case of California, tax rates
increased by 770% between 1989 and 1999, while cigarette sales declined by more than
60% between 1989 and 2010, and tobacco excise revenues increased from US$250 mil-
lion before 1989 up to US$845 million in 2010 (Chaloupka et al. 2012). In Brazil, the excise
tax amount per cigarette pack was increased in real terms by 81.4% between 2011 and
2015, while total cigarettes sales decreased by 35% in the same period. However, real
excise revenues in 2015 were still 17% higher than in 2011 (Iglesias, 2016).
2.4 Employment Impact of Tobacco Taxation
Despite its effectiveness in reducing tobacco consumption and increasing tax rev-
enues, there is often a debate over the effects of tobacco tax employment in the
tobacco industry. The tobacco industry generates jobs in diverse parts of the economy,
including farming, manufacturing, and wholesale sectors. However, it is important to
differentiate the employment that is from core-tobacco sectors (directly dependent on
tobacco production, such as farming and manufacturing) compared to tobacco-related
employment (jobs that are just partially dependent on tobacco, such as retail). Historic
analyses of the tobacco industry show that the industry has significantly reduced employ-
ment because, over time, the industry has become more capital intensive and farming
has become more efficient, so that job losses have occurred even in the absence of
tobacco control measures (NCI-WHO, 2017). The tobacco industry has sponsored studies
19The Economics of Tobacco Taxation and Employment in Indonesia to document the employment contribution of the sector. Usually these studies argue that tobacco control measures, such as higher taxation, would result in job losses in the tobacco industry and, consequently, increase unemployment (Zhang, 2002). In contrast, most academic studies have shown that tobacco control policies, such as taxation, have an overall neutral or positive impact on employment. Zhang (2002) argues that the (industry-sponsored) studies use unrealistic assumptions about impacts of sales drop, overestimate the number of jobs associated with the tobacco industry and, consequently, overestimate the possible negative impact of tobacco control measures on overall employment. However, the main problem is that those studies do not consider: (i) the expansionary employment effect of consumption substitution of smokers who redirect their expenditure toward other products after the tax increase—tobacco expendi- tures do not disappear from the economy; rather, they are redistributed to the consump- tion and production of other goods and services; and (ii) the expansionary employment effect of higher public expenditure after the tax increase.4 Studies simulating the impacts of tobacco control policies on employment depend on key assumptions. For example, studies applying input–output models first estimate the change in final consumer demand for goods and services resulting from a tobacco control policy (US NCI and WHO, 2016). The change in demand is composed of two com- ponents: (i) the reduction of tobacco consumption, and (ii) the expansion of expenditures in other products, according to consumers (smokers) preferences. These studies then calculate the induced changes in outputs based on input–output tables that describe the flow of goods and services within the economy. Finally, changes in outputs are converted into changes in employment to obtain the employment impacts. The critical assump- tions these studies rely on are: the impact of tobacco control measures—estimated price and income elasticities in the case of tax increases; and, more importantly, the type of consumption substitution that smokers display after the tax increase or tobacco control measure. The normal assumption is that ex-smokers would follow the average expendi- ture pattern or the most recent quitter expenditure pattern. If other selected goods and services have a larger direct and indirect employment input than tobacco product production, the net employment effect is positive. The challenge to measure the employment impacts of higher tobacco taxes is that employment losses could be relatively concentrated, whereas employment gains tend to spread throughout the economy. Table 1 presents a selection of recent independent studies in low- and middle-income countries (LMICs). Under the assump- tions discussed above, four out of the six studies resulted in net employment gains after 4 This expansionary effect depends on fiscal policy decisions; greater public revenues do not mean automatically higher public expenditures. 20 // The Economics of Tobacco Taxation
Table 1: Selected studies on net employment impact of tobacco control policies
STUDIES MODEL AND ASSUMPTIONS CONCLUSIONS
South Africa Static input–output model Net gain of 50,236
jobs occurred in 1995
Van der Merwe Domestic consumption expenditures were eliminated, and the rate of consumption
by eliminating tobac-
and Abedian, decline in 1995 doubled.
co expenditures, with
1999 Expenditures were allocated by recent quitter and average expenditure pattern. consumers acting as
recent quitters and with
the same government
Government spending was reduced or kept at the same level by increasing other taxes.
spending
Zimbabwe Static input–output model
Van der Merwe, Domestic consumption expenditures and tobacco production in 1980 were eliminated.
Net loss of 87,798 jobs
1998 in 1980, and 47,463 jobs
Average input–output pattern changed, and all tobacco production was shifted to when all output went to
alternative agriculture products. alternative agriculture
products
Because of increases in other taxes, no change in government spending occurred.
Bangladesh Static input–output model
Van der Merwe, Domestic consumption expenditures and all tobacco production for tobacco products
1998 and bidis in 1994 were eliminated.
Net gain of 10,989,192
Average input–output pattern changed, and all tobacco production was shifted to jobs in 1994
alternative agriculture products.
Because of increases in other taxes, no change in government spending occurred.
Bulgaria Static input–output model
Petkova and Domestic consumption expenditures and tobacco production in 1999 were eliminated.
colleagues, Net loss of 5,567 jobs
2003 Average input–output pattern changed, and all tobacco production was shifted to in 1999
alternative agriculture products.
Because of increases in other taxes, no change in government spending occurred.
Egypt Static input–output model
Nassar and Domestic consumption expenditures and tobacco production in 1999 were eliminated.
Metwally, 2003 Net loss of 5,567 jobs
Average input–output pattern changed, and all tobacco production was shifted to in 1999
alternative agriculture products.
Because of increases in other taxes, no change in government spending occurred.
Indonesia Static input–output model Net gain of 84,340 jobs
with a 25% tax increase;
Ahsan and Percentage increases of 25%, 50%, and 100% occurred in the cigarette tax. net gain of 140,567 jobs
Wiyono, 2007 with a 50% tax increase;
Expenditures were allocated by the average expenditure pattern.
and net gain of 281,135
jobs with a 100% tax
increase
Source: National Cancer Institute and World Health Organization (2017), Table 15.3, page 560.
21The Economics of Tobacco Taxation and Employment in Indonesia
tobacco control policies. Generally, the net effects were not significant, except in the case
of Bangladesh and Egypt. Reductions occurred in core tobacco sectors, including tobacco
farming and manufacturing; in tobacco-related sectors, such as wholesaling and retailing;
and in ancillary sectors, such as the paper and pesticide industries. The net gains
in employment depended on several factors, such as: the assumed structure of
population (smoker) consumption; the production structure of the economy, i.e., the
extent to which final products, inputs, and services were produced domestically or
imported in the tobacco industry and in the industries where ex-smokers would spend
their money; and the labor intensity of tobacco growing/manufacturing versus the rest
of the industries composed of the average consumer expenditures. For example, Ahsan
and Wiyono (2007) found that, in Indonesia, the top five sectors that would experience
increased employment include rice, tea, coffee, sugarcane, and root crops, which have
higher labor intensity than tobacco growing (Ahsan and Wiyono, 2007)
Table 2: Type of external tobacco trade and likely net employment impacts
TYPE OF EXTERNAL MEANING LIKELY EMPLOYMENT IMPACTS
TOBACCO TRADE THAT A
COUNTRY HAS
Net exporter of tobacco Production of tobacco leaf and Domestic tobacco demand is not the only determinant
products cigarettes is higher than domestic of production/employment in the core sectors.
consumption Drop of domestic tobacco demand could be compen-
sated with production directed to external markets,
Tobacco employment distribution and employment effects could be minimized.
in domestic sales or exports would The relative effects of global and domestic policies
depend on the share of exports to depends on the share of production that is exported.
total sales
Balanced tobacco economy Domestic production of tobacco Domestic tobacco control policies may have negative
leaf or cigarettes is used primarily net employment effects. Smoking prevalence and
for local consumption. Self-suffi- size of control policy could be important for absolute
cient in tobacco impact on net employment.
Net importers of tobacco Produce less tobacco leaf or ciga- Tobacco control policies in countries with small
products rettes than they consume tobacco/cigarette production may increase overall
employment. Not affected by global tobacco demand.
Mixed tobacco economy Significant grower and producer, U.S. is an example: tobacco leaf producer, importer
and imports and/or exports a sub- and exporter of large amounts of tobacco leaf.
stantial share of tobacco leaf and Changes in both domestic and global tobacco control
tobacco products policies would affect employment.
Source: Own elaboration based on US NCI and WHO (2016).
22 // The Economics of Tobacco TaxationThe possible employment impacts varied depending on whether the country was
a net exporter or net importer of tobacco leaf. When tobacco control policies reduced
the demand for cigarettes, a country was likely to have lower employment losses if that
country imported a significant percentage of the cigarettes smoked and/or leaf used
to make them, and domestically produced a large portion of the rest of the products
included in the average consumption expenditure. Conversely, the more the tobacco
leaf and other inputs and cigarettes were nationally grown and/or produced relative to
the local content of the things people buy instead, the greater is the likelihood that there
was some employment losses locally. The higher the production diversification of the
economy, the higher was the domestic employment created by the demand switch from
tobacco toward other products. In contrast, the higher the expenditure on tobacco and
the lower the sectoral diversification of domestic production, the smaller (or even neg-
ative) was the net employment impact. The United States National Cancer Institute (US
NCI) and the WHO published a report that presented a classification of countries and the
likely employment impacts, based on a country’s situation in terms of tobacco trade: net
exporter, net importer, balanced economy and mixed situation (US NCI and WHO, 2016).
As shown in Table 2, domestic tobacco control policies likely had a larger impact in coun-
tries with a balanced tobacco economy—self-sufficient in tobacco, because the employ-
ment destruction reduction of tobacco demand could not be offset.
Decline in tobacco consumption as a result of taxation or tobacco control policies
may occur gradually. Although most of the studies presented in Table 1 assume a sharp
and total reduction in cigarette consumption, smoking prevalence reduction occurred
gradually even when a significant tax increase was implemented in the short term.5
Tobacco control policies normally gradually reduce smoking, distributing adjustment
costs through time and diluting them in decades (US NCI and WHO, 2016). The adjust-
ment costs of labor-intensive segments of the industry, such as tobacco farming, depend
on the existence of viable alternatives. In many LMICs, such as Brazil and the Philippines,
tobacco farmers are diversified, producing other crops, which may facilitate the transition
to other crops. The result from the World Bank/ACS survey on the economics of tobacco
farming unequivocally demonstrated that former tobacco farmers were growing many
of the same crops as current tobacco farmers, and were simply increasing production of
these crops. Moreover, these former tobacco farmers were typically generating higher
revenues and incurring lower costs than their neighbors and peers who continued to
grow tobacco.
5 The Philippines, for example, increased the excise taxes for the cheapest cigarettes by 341% in a one-year period that resulted in a
decline of tobacco use prevalence by approximately 20% Global Adult Tobacco Survey (GATS).
23The Economics of Tobacco Taxation and Employment in Indonesia The employment issue has lost preeminence in the debate of tobacco control policies in many countries. Since the first half of the last decade, there have been few studies about the employment effect of tobacco control policies from either side of the debate; instead the debate has concentrated on other issues, such as illicit trade in tobacco products. For example, in the recent Brazilian experience of large increases of tobacco taxes (2011–2016), effects on employment were not an issue, because the export share of leaf production in the country increased. In the Philippines (2012 onward) poten- tial employment impact was a big issue during Congressional deliberations, but since 80% of tobacco leaf was exported, the Ministry of Finance recognized that the employ- ment impact risk on tobacco farmers was low. Furthermore, a budget was also allocated to help those that may be adversely affected—15% of the incremental tobacco tax reve- nues was allocated to tobacco farmers to encourage them to shift to alternative crops. 24 // The Economics of Tobacco Taxation
25
SMOKING HAS BEEN A MAJOR CONTRIBUTOR TO THE DISEASE BUR IN INDONESIA. SMOKING PREVALENCE AM WORKING-AGE INDIVIDUALS HA EXCEEDED 30
3
CIGARETTE TAX POLICIES IN
INDONESIA
Indonesia has one of the most complex cigarette excise tax structures in the world.
The current cigarette excise tax system has 12 tiers, which are based on manufacturers’
types of cigarettes, the scale of cigarette productions, and per unit retail price.6 There are
three types of cigarettes: machine-made kreteks (SKM), machine-made white cigarettes
(SPM), and hand-rolled kreteks (SKT).7 Manufacturers of either SKM or SPM are considered
as a Class I if they produce more than 3 billion cigarettes annually and as a Class II if they
produce less than the 3 billion cigarettes. On the other hand, manufacturers of SKT are
considered a Class I if they produce more than 2 billion cigarettes annually, a Class II if
they produce between 500 million to 2 billion cigarettes annually, and a Class III if they
produce less than 500 million cigarettes annually.
Excise taxes for cigarettes produced by larger manufacturers increased at a modest
rate between 2010 and 2017 (Figure 1). As shown in column 4 of Table 3, the excise
taxes for machine-made kreteks (SKM, Class I) and white cigarettes (SPM, Class I) increased
by 27 and 46 percent, respectively in real terms between 2010 and 2017. Note that
changes in the tariffs for the small-scale manufacturers were relatively smaller compared
to changes for the Class I SKM and SPM manufacturers. For example, excise taxes for Class
II SKM manufacturers increased 24 percent while excise taxes for Class II SPM manufactur-
ers increased 35 percent. It is important to note that tax increases that occurred have not
been enough to reduce cigarette affordability. The US NCI and WHO show that cigarettes
in Indonesia were much more affordable in 2013 than they were in 2000 (US NCI and
WHO, 2016).
The tiers in the cigarette excise tax structure aimed to accommodate small-scale
cigarette firms, especially SKT firms. The rationale for such a structure was to protect
smaller SKT firms that accounted for more than half of total factories in the tobacco indus-
try (column 5 of Table 3). Moreover, these firms were responsible for employing a signifi-
cant share of the workers in tobacco manufacturing. For example, the per-unit tariffs for a
SKM produced by Class I manufacturers and sold for a minimum of IDR 1,120 per unit was
IDR 530 (47.3 percent of retail sale price), while the per-unit tariffs for a SKT produced by
6 Ministry of Finance Decree Number 43/PMK.04/2005.
7 In addition to SKM, SPM, and SKT, there are hand-rolled white cigarettes with filters. The structure of the excise tax for these products
are identical to the structure for SKM..
27The Economics of Tobacco Taxation and Employment in Indonesia
Table 3: Cigarette excise tax in Indonesia, 2017
TYPE OF CATEGORY PER-UNIT % NOMINAL NUMBER
CIGARETTE (PRODUCTION) TARIFFS CHANGE OF
(TIERS), IN (REAL), FACTORIES
RUPIAH 2010-2017 (2015)
Machine-made kreteks (SKM), hand-rolled Class I, > 3 billion 530 78 (27) 14
kreteks with filter (SKTF), hand-rolled white sticks
cigarettes with filter (SPTF)
Hand-rolled white cigarettes with filter (SPTF) Class II, ≤ 3 billion 335-365 75 (24) 232
sticks
Machine-made white cigarettes (SPM) Class I, > 3 billion 555 106 (46) 1
sticks
Class II, ≤ 3 billion 290-330 90 (35) 25
sticks
Hand-rolled kreteks (SKT) Class I, > 2 billion 265-345 66 (18) 16
sticks
Class II, 500 million–2 155-165 64 (15) 21
billion sticks
Class III, ≤ 500 million 80-100 38 (-2) 404
sticks
Source: Peraturan Menteri Keuangan (Ministry of Finance Decree) abbreviated as PMK: PMK Number 181 2009,
PMK Number 190 2010, PMK Number 167 2011, PMK Number 205 2014, PMK Number 198 2015, PMK Number
147 2016.
Note: For parsimony, per-unit tariffs for Class II and Class III categories are simplified from per-unit tariffs for Class
IIA, Class IIB, Class IIIA, and Class IIIB. Real change in per-unit tariffs accounts for changes in prices of goods and
services in the economy.
Class I manufacturers and sold for a minimum of IDR 1,215 per unit was only IDR 345 (28.4
percent of retail sale price). These numbers imply that per-unit tariffs for SKT was half of
the per-unit tariffs for SKM of a similar tier. The ratio between the highest and the lowest
per-unit tariffs in the 2017 tax structure was quite high at 6.7. Changes in real excise tax
for SKT between 2010 and 2017 was also quite low at 18% or lower. Additionally, SKT pro-
duced by Class III producers were cheaper, in real terms, in 2017 than they were in 2010.
Despite the low tax rates, there was a downward trend in the sale of SKT in the
market. In 2001, the share of SKT in the market was about 40% (Ministry of Industry,
2009). The share decreased to 35.5% in 2005, to 30% in 2011, and to just 26% in 2013. We
argue that the shift away from SKT was not likely the result of higher cigarette taxes. 8
First, as shown in Table 3, taxes imposed on SKT were among the lowest in the cigarette
8 Particularly because, as noted above, the tax increases were not enough to reduce affordability. cigarettes in Indonesia were much
more affordable in 2013 than they were in 2000 (US NCI and WHO, 2016).
28 // Cigarette Tax Policies in IndonesiaFigure 1. Trends of the Cigarette Excise Tax in Indonesia in real terms, 2010–2017
EXCISE TARIFF PER STICK, IN REAL TERMS
400 400
300 300
200 200
100 100
2010 2011 2012 2013 2014 2015 2016 2017 2010 2011 2012 2013 2014 2015 2016 2017
HRK Class I, layer 1 HRK Class I, layer 2 HRK Class II, layer 1 HRK Class II, layer 2
MMK Class I MMW Class I MMK Class II, layer 1 MMK Class II, layer 2
MMW Class II, layer 1 MMW Class II, layer 2
Note: the pre - 2009 tariffs are calculated based on the current 12 layers
Source: : Peraturan Menteri Keuangan (PMK) 181 2009, PMK 190 2010, PMK 167 2011, PMK 179 2012, PMK 205 2014, PMK 198 2015, PMK 147 2016
Source: Peraturan Menteri Keuangan (Ministry of Finance Decree) abbreviated as PMK: PMK Number 181 2009,
PMK Number 190 2010, PMK Number 167 2011, PMK Number 205 2014, PMK Number 198 2015, PMK Number
147 2016.
Note: The pre-2009 tariffs were calculated based on the current 12 layers.
industry, and the changes of the taxes in real terms were quite small. Second, despite the
decreasing share, productions of SKT were still increasing over time at least up to year
2010 (Tobacco Control Support Center, 2014). The shift away from SKT have been driven
by changing preferences among smokers for machine-made products due to income
growth. It may also have reflected substitution of more capital-intensive technology away
from labor-intensive hand-rolling in cigarette production.
3.1 Effects on Cigarette Consumption, Revenues, and Employment
Previous evidence suggests that raising cigarette prices through taxes may have
potential to reduce consumption in Indonesia. A review by Setyonaluri and col-
leagues (2008) reported price elasticities of demand ranging from –0.26 to –0.76. These
estimates suggest that raising prices through taxes can reduce consumption. Ahsan
(2011) suggested that a 16 percent increase in cigarette tax would reduce consumption
by 4.7 percent, implying a price elasticity of –0.29. The impacts of raising cigarette tax on
health were also potentially significant. A study suggested that raising cigarette taxes to
50 percent of retail prices can reduce expected mortality rate by 2 to 5 percent, depend-
ing on the price elasticity of demand (Setyonaluri et al., 2008).
29The Economics of Tobacco Taxation and Employment in Indonesia This report updates earlier elasticity estimates. To do so, we used the 2015 National Socioeconomic Survey (Susenas) to estimate the price elasticity of cigarette demand.9 The NSS contains data on whether an individual smoked kreteks or white cigarettes, the number of cigarettes smoked in the past week, and total expenditure for cigarettes. The limitation of our estimation was that we could not differentiate between hand-made and machine-made kretek. However, obtaining separate estimates for kreteks and white cigarettes offered a significant improvement over the existing literature. This information allowed estimating price elasticity of demand for SKT and SKM separately. Results showed variation in price elasticities of demand between kreteks and white cigarettes. The esti- mated price elasticity of demand for kreteks was about –0.42 while for white cigarettes was about –0.51 (within the range of –0.26 and –0.76 reported in previous studies). We acknowledge our price elasticities may be underestimated because they do not account for substitutions to cheaper products. For example, individuals can easily switch to cheaper cigarettes if cigarette prices increase because of higher excise taxes. An important aspect is to observe the impact of tobacco excise tax across income groups. The global evidence suggests that cigarette tax was regressive and showed that the young and the poor were more responsive to price changes. For example, Nasrudin and colleagues (2013) estimated price elasticities of –0.15 to –0.16 among individuals in the first two income deciles and price elasticities of –0.20 to –0.28 among individuals in the third to ninth deciles. The authors estimated that the tax burden as a percentage of income of individuals in the first two deciles was between 3 to 10.63 percent and the tax burden of individuals from the other income groups was below 2 percent (Nasrudin et al., 2013). These results may be explained by the policy of keeping kreteks cheap. However, the study did not consider the benefits of raising cigarette taxes. A more recent study showed that once indirect benefits (such as lower health care expenditures and higher productivity) of raising cigarette taxes were accounted for, a tobacco tax increase was actually progressive (Fuchs and Meneses, 2017). In Indonesia, annual excise tax revenues from tobacco products had been increas- ing monotonically since 2005 (Figure 2). In 2007, the realized excise tax revenue was IDR 43.54 trillion (US$4.87 billion). The figure grew to IDR 55.38 trillion (US$5.33 billion) in 2009, IDR 103.6 trillion (US$9.90 billion) in 2013, and IDR 139.5 trillion (US$11.76 billion) in 2014 (IAKMI, 2014; Dwi Kurnaini, 2016).10 Historically, tobacco excise accounts for more than 95 percent of total excise revenue. Simulation studies suggest that a 10 percent increase in cigarette tax would increase excise revenues by 6.7 to 9.0 percent (Setyonaluri et al., 2008). Another simulation suggests that a 16 percent increase in cigarette tax would 9 We report the estimations of price elasticity of demand in Annex III. 10 The exchange rates were IDR 8,938/US$ in 2004, IDR 10,389/US$ in 2009, IDR 10,461/US$ in 2013, and IDR 11,865/US$ in 2014 (OECD, 2017). 30 // Cigarette Tax Policies in Indonesia
You can also read
